301 Been Possible Due to Restricted Access

301 Lockley, M.G. & Lucas, S.G., eds., 2014, Fossil footprints of western North America: NMMNHS Bulletin 62

NOTES ON A NEW ANKYLOSAUR TRACK FROM THE DAKOTA GROUP (CRETACEOUS) OF NORTHERN COLORADO

MARTIN G. LOCKLEY 1 and GERARD D. GIERLINSKI 2

1 Dinosaur Tracks Museum, University of Colorado at Denver, PO Box 173364, Denver, Colorado, 80217; e-mail: [email protected]; 2 JuraPark, ul. Sandomierska 4, 27-400 Ostrowiec Swiztokrzyski, Polish Geological Institute, ul. Rakowiecka 4, 00-975 Warszawa, Poland; Poland; e-mail: [email protected]

Abstract—A new report of an isolated, but well-preserved ornithischian track from a not-in-situ locality in northern Colorado is confidently attributed to an ankylosaurid trackmaker from the Dakota Group. It is one of the largest and oldest tracks yet recovered from the Dakota Group, and appears to have originated from the lower part of the section, where tracks are typically rare.

INTRODUCTION University of Colorado Natural History Museum; YPM, Yale Peabody Museum in New Haven, Connecticut Many tracks preserved as large natural casts have been reported from Cretaceous siliciclastic facies in the Western Inte- GEOLOGICAL PROVENANCE rior region of the western USA. Reports of such occurrences in the Dakota Group are almost too numerous to mention, as The track was found in a gully on private land (from which evidenced from a number of papers in this volume (and refer- it was collected in 2009) at a site about 33 km north of Fort ences therein). Likewise, many reports of similar occurrences Collins, Colorado, before being donated to the University of are associated with the “MesaVerde Group,” especially in coal Colorado Denver Dinosaur Tracks Museum. According to avail- mines (Parker and Balsley, 1989; Parker and Rowley, 1989). able geological maps (Courtright and Braddock, 1989) the track These track-rich deposits, loosely referred to as coal-bearing, was found sitting on outcrops designated as belonging to the coastal-plain facies (Lockley et al., 1994), consist of heterolithic Triassic Lykins Formation (Fig. 1). Analysis of this map shows (anisotropic) sequences of sandstone and mudstone, that appear that the discovery site lies at the base of a slope capped by units to have been very conducive to the preservation of sandstone of the Cretaceous Dakota Group (Fig. 1). The lower part of the casts, especially in the coastal plain systems associated with the slope designated as (TRlu) represents the Triassic upper Lykins Western Interior Seaway. Formation, which is overlain, in ascending order by the follow- Many of these casts represent the sandy infillings of deep ing mapped units: the undivided Triassic Jelm and Middle-Up- tracks made in muddy substrates by large dinosaurs and, as a per Jurassic Sundance formations (designated as JTRsj), the result, once they lithified, they become very large, heavy and Jurassic Morrison Formation (Jm) and the Lytle Formation and bulbous natural casts. Once they weather out of outcrops they Plainview Sandstone Member of the South Platte Formation slide and roll down slopes, and are frequently found out of place: (Kspl): see Fig.1. As noted below (Fig. 1), the nomenclature and i.e., no longer in situ. Depending on the size and depth of these inferred age of pre-Morrison units is of no direct interest in casts they may weigh anywhere from ~5 to ~200 kg. While this determining the provenance of the track described here, and for may make collecting of the smaller, less-heavy casts easy, deter- this reason they are not discussed further. mining the horizon(s) from which they originated may be diffi- Based on the known distribution of large ornithischian cult, especially in cases where there are multiple track-bearing tracks it seems impossible that the track could have originated layers in similar lithologies. from anywhere below the Morrison Formation, and it is more We report on a single, large natural cast of a track weighing likely to have originated from the Dakota Group. Based on the ~75 kg that was discovered as a loose specimen from near Fort topography (see the cross section shown in Fig. 1) it is most Collins, Colorado (Fig. 1), and donated to the CU Denver Dino- probable that the track was eroded from the aforementioned saur Tracks Museum in 2009. The specimen (UCM 209.149, lower mapped unit of the Dakota Group (Kspl). If the track had formerly CU 29.149; Fig. 2) is a tetradactyl pes cast attributed to been eroded from one of the higher units in the Dakota Group, an ornithischian, specifically an ankylosaur, trackmaker. As the such as the Middle Shale member (Ksm) or First Sandstone locality is under private ownership, revisiting the site to locate member (Ksf) exposed farther to the northeast, barring the pos- the track-bearing layer and look for other specimens has not sibility of glacial transport, it appears that it would have been been possible due to restricted access. impossible to have been transported to the location where it was found without having had to have moved “up” the Kspl dip INSTITUTIONAL ABBREVIATIONS slope! Assuming an origination point in the Morrison Formation or Dakota Group, it is possible to compare this track with other CEUM, CEU Prehistoric Museum in Price, Utah; CU, Uni- ornithischian tracks found in these units. versity of Colorado at Denver, Colorado; CU-MWC, University This report is of interest because a smaller tetradactyl track of Colorado and Museum of Western Colorado joint collection; was observed by the senior author at an in situ outcrop at the top MOR, Museum of the Rockies in Bozeman, Montana; UCM of the Lytle-Plainview Formation near Bellvue, Colorado about 302

FIGURE 1. Locality map showing ankylosaur track locality discussed in this paper. Top left: location of site in northern Colorado. Bottom left: stratigraphic units after Courtright and Braddock (1989). See text for details. Top right and lower right: map and corresponding cross section showing where track was found at base of slope capped by Cretaceous rocks. See text for details.

10 km northwest of Fort Collins. The stratigraphy of this locality the shortest, and a digital length formula of II=III>IV>I. Measured is well known and sections have been published by Macke and between the hypicies, the toe casts are more or less equal in width Maughan (1985) and Weimer et al. (1990) (~11-12 cm wide), with those of digits II and III being longer (~17- 18 cm from hypex to toe tip) than digit IV (~ 14 cm) and digit I TRACK DESCRIPTION (~9-10 cm). Thus, the longest toes measured between toe tips and the hypicies on either side of digit III represent about 31-32% of The track cast (UCM 209.149) is of a tetradactyl ornithischian the overall length of the cast. The distal part of the cast of toe II right pes with wide, blunt toes (Fig. 2). The cast is 57 cm long and shows what appears to be the cast of an ungual protruding from 48 cm wide, and is broadly triangular in shape with the acute apex the cast of the fleshy part of the foot that surrounds it. Such traces, represented by the heel. As the posterior part of the heel trace may which appear to differentiate distinct morphological features of the represent forward and downward movement into the substrate as foot, are rare but have been reported in the case of ornithopods the foot registered, these raw cast measurements may not exactly (Lockley et al., 2006a, fig. 19) and theropods (Huerta Hurtado, correspond to the original foot size. However, because the track is 2001; Torcida Fernandez-Baldor et al., 2007). well-preserved, these measurements probably represent foot size with fair accuracy. The cast indicates a footprint at least 30 cm DISCUSSION deep. However, because there are no indications on the cast as to where the original substrate surface may have been relative to the We divide the following discussion into two parts; the first deepest part of the track, the depth estimate is imprecise. Neverthe- dealing with the age and occurrence of presumed ankylosaur less, as noted below, it is clear that the trackmaker was a large indi- tracks, and the second dealing with their morphological character- vidual in comparison with any other reported from this time interval. istics and implications for trackmaker identity. The toe casts are inferred to represent digits I-IV with digit I 303

FIGURE 2. The reported specimen: cf. Tetrapodosaurus sp. (UCM 209.149, formerly CU 209.149), presumed to be from the Dakota Group of northern Colorado. Photograph (left) shows natural cast of presumed right footprint. Line drawing (right) is a reversed tracing showing presumed ungual traces on digits II and IV.

Age and Occurrence of Dakota Group Ankylosaur Tracks Morphological Characteristics

Although ornithischian dinosaur tracks are common in the It is difficult to infer the identity or taxonomic affinity of track Dakota Group, especially in the upper part (Sequence 3 sensu makers responsible for making tracks and trackways that appear Weimer, 1989; Lockley et al., 1992, 2006b), most of these tracks attributable to the broad category of quadrupedal ornithischian. In have been attributed to ornithopods (ichnogenus Caririchnium). a recent review of all ornithischian track morphotypes (Lockley Ankylosaur tracks are relatively rare in the Dakota Group of east- et al., 2012), six categories were recognized, including those of ern Colorado. The first significant report (Kurtz et al., 2011) dealt small basal ornithischians with tridactyl pes footprints and large with an assemblage of large tracks inferred to be from Sequence ornithopods (iguanodontids and hadrosaurs), also with tridactyl 2 (Plainview Formation) near Cañon City, Colorado. This report pes footprints. In this discussion we deal only with trackmakers documents large pes tracks up to 48 cm long and 52 cm wide. with a tetradactyl pes. This reduces the categories to four mor- The only other documented occurrences in eastern Colorado deal photypes, most of which are of ornithischian affinity. However, as with specimens from Sequence 3 that are much smaller (Lock- noted below, one category, containing the ichnogenus Marcropo- ley et al., 2006) and include specimens about 27.5- 32.0 cm long dosaurus, may be attributable to a non-ornithischian trackmaker. and 24.0-27.0 cm wide (Fig. 4g) that are more gracile than the The specimen UCM 209.149 described here (Fig. 2) is a clearly specimen reported herein. These are only about half the linear semiplantigrade footprint with four well-defined toes. Among the dimensions of the northern Colorado specimen (UCM 209.149) four main morphotypes (A-D) of semiplantigrade and plantigrade and the specimens from Skyline Drive (Kurtz et al., 2001). There- tracks from the Cretaceous of North America (Fig. 3), we refer fore, it is of stratigraphic significance that the only two reports UCM 209.149 to morphotype D. of well-preserved ornithischian tracks from the Dakota Group of eastern Colorado appear to be from the lower part of the Dakota Group (Sequence 2 and the lower part of Sequence 3). As reported Morphotype Categories elsewhere in this volume, ankylosaur tracks are common at some recently-found sites in western Colorado (Lockley et al., 2014), Morphotype “A” (Fig. 3A) has been recently recorded in but these are smaller than the Fort Collins and Skyline Drive speci- North America (Gierlinski and Lockley, in press) and represents an mens, and their relationship to the stratigraphic sequences of east- elongate footprint with four slender, tapering and narrowly divari- ern Colorado has yet to be determined. cated digits. This type closely resembles the Eurasian ichnotaxon named Macropodosaurus by Zhakarov (1964) and is identified as 304

FIGURE 3. Four main morphotypes of dinosaurian tetradactyl semiplantigrade and plantigrade tracks from the Cretaceous of North America: a - cf. Macropodosaurus sp. (CU 219.3) from the Ferron Sandstone at Muddy Creek Canyon in Utah; b - ceratopsian track (uncatologued specimen) from the Castlegate Sandstone at Thompson Pass in Utah; c - ceratopsian track (CEUM 831) from the Blackhawk Formation near Price in Utah; d - ceratopsian track (CU 227) from the Iron Springs Formation at Parowan Gap in Utah; e - Ceratopsipes goldenensis Lockley and Hunt 1995b (CU-MWC 220.516) from the Laramie Formation at PArfet Clay Pit in Colorado; f - Tetrapodosaurus sp. (CU-MWC 209.33) from the Dakota Group near Walsh in Colorado. belonging to a therizinosauroid track maker by Sennikov (2006). skeleton with shorter and wider phalanges, and metatarsals (Fig. This type differs importantly from specimen UCM 209.149 in 4F). However, Morphotype C is a wider form with shorter heel and that it possesses long sharp digit traces and a relatively narrow U- slightly shorter digits than observed in track UCM 209.149. shaped heel in comparison with other morphotypes. Morphotype “D” (Fig. 3F), previously also supposed to be Morphotype “B” (Fig. 3B-C) is characterized by a relatively of ceratopsian affinity (Sternberg, 1932; Gierlinski and Sabath, narrow metapodium, but one that is V- rather than U-shaped poste- 2008), seems to fit well the morphology of discussed specimen. riorly, and highly prominent (anteriorly projecting) traces of digits However, this type is represented by the ichnogenus Tetrapodo- II and III. This kind of tetradactyl footprint is mainly known from saurus (Sternberg, 1932), is more likely attributed to an ankylo- the Upper Cretaceous coal mines near Price, Utah, and is generally saurid trackmaker (McCrea et al., 2001). The type of Tetrapodo- inferred to be of ceratopsian origin (e.g., Parker and Rowley, 1989; saurus has rather elongate and well-demarcated pes digit traces, Lockley and Hunt, 1995b). Its morphology fits a gracile ceratop- in comparison with Ceratopsipes (Morphotype C). However, we sian foot pattern, possibly corresponding to the pedal morphology recognize that the amount of flesh contained by and the separation of Cerasinops (Fig. 4C). This is also clearly a different type of of the pes digits and their corresponding traces may be variable morphology from specimen UCM 209.149, which shows very and size related. Thus, large ankylosaur pes feet and tracks may short broad toes subequal in length. be convergent with those of large ceratopsians. Conversely, there Morphotype “C” (Fig. 3D-E) comprises ichnites such as may also be convergence between the tracks and feet of the smaller Ceratopsipes (Lockley and Hunt, 1995b) and an unnamed ceratop- members of these two clades. sian track from the Iron Springs Formation of Utah (Milner et al., 2006), identified as the footprints of more derived ceratopsians. This morphotype corresponds with a more robust ceratopsian foot 305

FIGURE 4. Three morphotypes of tetradactyl semiplantigrade and plantigrade ornithischian tracks in comparison with the foot skeletons of possible trackmakers: a - ceratopsian track (un-cataloged CEUM specimen) from the Castlegate Sandstone of Utah; b - ceratopsian track (CEUM 831) from the Blackhawk Formation of Utah; c - Cerasinops pes (MOR 300) from the Two Medicine Formation of Montana; d - ceratopsian track (CU 227) from the Iron Springs Formation of Utah; e - Ceratopsipes goldenensis Lockley & Hunt 1995b (CU-MWC 220.516) from the Laramie Formation of Colorado,; f - Centrosaurus pes (AMHN 5351) from the Dinosaur Park Formation of Alberta; g -Tetrapodosaurus sp. (CU-MWC 209.33) from the Dakota Group of Colorado; h – studied track CU 209 from the Platte Forma- tion of Colorado; i - ankylosaur pes reconstructed on the basis of an Nododosaurus material YPM 1815 from the Frontier Formation of Wyoming. 306 CONCLUSIONS Lockley, M.G., Holbrook, J. Kukihara, R., and Matsukawa, M., 2006b, An ankylosaur-dominated dinosaur tracksite in the Cretaceous Dakota Group of Colo- rado and its paleoenvironmental and sequence stratigraphic context: New The study of the tracks of ankylosaurs and related quadrupe- Mexico Museum of Natural History and Science, Bulletin, 35, p. 95-104. dal ornisthischians is still in flux (Lockley et al., 2012), and has Lockley, M.G., Hunt, A.P., and Meyer, C., 1994, Vertebrate tracks and the ichnofa- been somewhat hampered by the lack of well-preserved and well- cies concept: Implications for paleoecology and palichnostratigraphy; in defined trackways. Although abundant trackways are known from Donovan, S., ed., The Paleobiology of Trace Fossils: Wiley and Sons, Inc., p. some regions (McCrea et al., 2001, 2014), in many regions the 241-268. sample of ankylosaur tracks is mainly composed of isolated casts. Lockley, M.G. and Hunt, A.P., 1995a, Dinosaur tracks and other fossil footprints of the western United States. Columbia University Press. New York, 338 p. The present study provides a typical example of an intriguing foot- Lockley, M.G. and Hunt, A.P., 1995b, Ceratopsid tracks and associated ichnofauna print with a distinctive morphology that has an uncertain geologi- from the Laramie Formation (Upper Cretaceous: Maastrichtian) of Colorado: cal provenance. Journal of Vertebrate Paleontology, v. 15, p. 592–614. Our tentative conclusion is that the track can be assigned to Lockley, M.G., Houck, K. Yang, S-Y., Matsukawa. M and Lim, S-K., 2006a, ichnogenus Tetrapodosaurus, herein assigned to Morphotype D. Dinosaur dominated footprint assemblages from the Cretaceous Jindong This inference suggests an ankylosaur track maker. The track Formation, Hallayo Haesang National Park, Goseong County, South Korea: Evidence and implications: Cretaceous Research, v. 27, p. 70-101. likely originated from the lower part of the Dakota Group, locally Macke, D.L. and Maughan, E.K., 1985, Rocky Mountain Section Field Trip Guide. mapped as the Lytle Formation and Plainview Sandstone Member Rocky Mountain Section America Association of Petroleum Geologists and of the South Platte Formation Lytle-Plainview (Fig. 1). This con- Rocky Mountain Section Society of Economic Paleontologists and Mineralo- clusion is of interest because previous reports of tracks from this gists. unit are rare despite an abundance of reports from the upper part McCrea, R., Lockley, M.G. and Meyer, C.A., 2001, Global distribution of pur- of the Dakota Group. ported Ankylosaur track occurrences; in Carpenter, K., ed., The Armored Dinosaurs. Indiana University Press. p. 413-454. McCrea, R.T., Buckley, L.G., Plint, A.G., Currie, P.J., Helm, C.W., Haggart, J.W., ACKNOWLEDGMENTS and Pemberton, S.G., 2104, A review of vertebrate track-bearing formations from the Mesozoic and earliest Cenozoic of western Canada with a descrip- We thank Denis Anderson of Fort Collins for collecting tion of a new theropod ichnospecies and reassignment of an avian ichnogenus: New Mexico Museum of Natural History and Science, Bulletin 62, this vol- the specimen (CU 209.149) and donating it to the University of ume. Colorado Denver Dinosaur Tracks Museum. Richard McCrea and Milner, A.R.C., Vice, G.S., Harris, J.D. and Lockley, M.G., 2006, Dinosaur tracks Lisa Buckley, Peace Region Palaeontological Research Center, from the Upper Cretaceous Iron Springs Formation, Iron County, Utah: New provided helpful reviews of this manuscript. Mexico Museum of Natural History and Science, Bulletin 35, p. 105-113. Parker, L.R. and Rowley, R.L. Jr., 1989, Dinosaur footprints from a coal mine in REFERENCES east-central Utah; in Gillette, D.D. and Lockley M.G., eds., Dinosaur Tracks and Traces. Cambridge University Press, Cambridge, p. 361-366. Courtright, T.B. and Braddock, W.A., 1989, Geologic map of the Table Mountain Parker, L.R. and Balsley, J., 1989, Coal mines as localities for studying dinosaur Quadrangle and adjacent parts of the Round Butte and Buckeye Quadrangles, trace fossils; in Gillette, D.D. and Lockley M.G., eds., Dinosaur Tracks and Larimer County Colorado, and Larimer County Wyoming: U. S. Geological Traces. Cambridge University Press, Cambridge, p. 353-359. Survey, Miscellaneous Investigation Series Map 1-1805. Sennikov, A.G., 2006, Reading segnosaur tracks: Priroda, v. 5, p. 58-67 [in Russian]. Gierlinski, G. and Lockley M. G. 2013. First report of probable therizinosaur (cf. Sternberg, C.M., 1932, Dinosaur tracks from Peace River, British Columbia: Macropodosaurus) tracks from North America, with notes on the neglected Bulletin of the National Museum of Canada, v. 68, p. 59–85. vertebrate ichnofauna of the Ferron Sandstone. Chapter 23, In Titus, A. L. and Torcida Fernandez-Baldor, F.T., Izquierdo Montero, Huerta, P. Montero Huerta, D., Loewen, M. A. (eds). At the Top of the Grand Staircase, Indiana University Perez Martinez, G and Urien Montero, V., 2006, El Yacimiento de icnitas de Press, p. 530-535. dinosaurios de Costalomo (Salas de los Infantes, Burgos, España): Nuevos Gierlinski, G. and Sabath, K., 2008, Stegosaurian footprints from the Morrison datos: III Jornadas Internacionales sobre Paleontologia de Dinosaurios y su Formation of Utah and their implications for interpreting other ornithischian Entorno, Salas de los Infantes, Burgos, p. 313-347. tracks: Oryctos, v. 8, p. 29-46. Weimer, R.J., 1989, Sequence stratigraphy, Lower Cretaceous, Denver Basin, Col- Huerta Hurtado, P., 2001, Procesos de formacion de huellas en Costalomo (Salas de orado U.S.A.; in Ginsburg, R.D. and Beaudoin, B., eds., Cretaceous Resourc- los Infantes, Burgos: Actas de las I Jornadas Internacionales sobre Paleonto- es, Events and Rhythms. NATO ASI Series, Kluwer, Dordrecht, p. 1-8. logia de Dinosaurios y su Entorno. Colectivo Archeologico-Paleontologico Weimer, R.J., Dolson, J., Etheridge, F. and Warme, J., 1990, Sequence Stratigra- de Salas, p. 361-369. phy-Lower K, Fort Collins-Boulder-Morrison Area. 2-Day Feld Trip (April Kurtz, B. Jr. Lockley, M.G. and Engard, D., 2001, Dinosaur tracks in the Plainview 28-29). Rocky Mountain Section of the Society for Sedimentary Geology Formation, Dakota Group (Cretaceous, Albian) near Cañon City, Colorado: (Society of Economic Paleontologists and Mineralogists), 125 p. A preliminary report on another “dinosaur ridge:” Mountain Geologist, v. 38, Zhakharov, S.A., 1964, On footprints of a Cenomanian dinosaur whose tracks were p. 155-164. found in the Shirkent River Valley; in Reiman, V. M., ed., Paleontology of Lockley, M.G., Cart, K. Martin, J., Prunty, R., Hups, K., Lim, J-D., Kim, K-S. Tadzikistan:Akademiya Nauk Tadzhikskoi SSR, Dushanbe, p. 31-5 [in Rus- Houck, K., Gierlinski, G. and Milner, A.R.C., 2014, A bonanza of new sian with English summary]. tetrapod tracksites from the Cretaceous Dakota Group, western Colorado: Implications for paleoecology: New Mexico Museum of Natural History and Science, Bulletin 62, this volume. Lockley M.G., Gierlinski, G. McCrea, R.T, and Buckley, L., 2012, Tracking thyreo- phorans and related ornithischian dinosaurs: a review; in Lida, X. and Lockley, M.G., eds., Abstract Book of Qijiang International Dinosaur Tracks Symposium, Chongqing Municipality, China Nov 29-30th 2012, Boulder Publishing, p. 62-63.